Populations with extreme phenotypes provide unique insights into the mechanisms of evolution. Island vertebrates are often unusually large or small compared to their mainland relatives. The repetition of this pattern across a diverse array of species raises the exciting possibility that common evolutionary and genetic mechanisms are responsible. Despite decades of debate about the evolutionary causes of this island rule, little information exists about its genetic basis. The largest wild house mice in the world reside on Gough Island; they have evolved extreme body size in just hundreds of generations. Gough Island mice belong to the same subspecies as the laboratory mice widely used in biomedical research, providing a powerful platform for the genetic dissection of body size evolution. Using this unique system, we will integrate genetic mapping, comparative morphometrics, and population genomics to provide the first detailed genetic portrait of rapid body size evolution in an island mammal. Specifically, we will (1) identify quantitative trait loci (QTL) responsible for the evolution of extreme body size and shape in Gough Island mice, and (2) find genomic regions associated with rapid adaptation to the island habitat using new genome sequences from this population. This research will reveal key characteristics of the genetic architecture of rapid evolution, including the number of loci and their phenotypic effects. By performing a series of intercrosses, we will simultaneously dissect the genetics of island-mainland differences and within-island variation. Recent selective sweeps in the Gough Island population will be identified using analyses of genome-wide patterns of variation that take into account the effects of non-equilibrium population history. Our combination of phenotype-driven and DNA-driven studies will address the contribution of natural selection to the island rule and has the potential to rapidly localize mutations responsible for th evolution of extreme body size. Several common human diseases, including cancer, obesity, diabetes, and heart problems, are directly connected to body size. Our genetic portrait of size variation in Gough Island mice will provide a comparative context for genome-wide association studies of size-related disease in humans. In addition, our results will help situate phenotypic differences among laboratory mouse strains used as models for disease within the larger context of natural trait variation.